This project was something I worked on when I worked at Memorial Sloan Kettering Cancer Center in the spring of 2023. The deliverable was a dataset in .csv format to be analyzed by the Epidemiology/Biostatistics team at MSK.
The files within this repository are as follows:

active-patient-cohort.sql
  This query is the start of the creation of the cohort of patients analyzed. It looks at the patients during the time period of 4/6/2022 to 5/6/2022 (6 months pre and post CURES deployment) who had any type of encounter in the 2 months before the week of interest. 

portal-active-patient-cohort.sql
  This query expands on the previous query and looks at if a patient is active on the patient portal during the specified weeks of interest. It is left joined on the active patient query.

There are additional queries that look at reports read, portal secure message (PSM) that are sent and the number of days where there's >= 1 login to either a patient or proxy account. All of these queries (6 total) were published to a Dremio data lake via Minio and python. They were then pulled into a seperate Jupyter notebook and agggregated.

CURES_etl_final_data.ipynb
  This python script creates and validates the final dataset to meet all the business logic for the biostats team.